Fakülteler / Faculties

Permanent URI for this communityhttps://hdl.handle.net/11727/1395

Browse

Filters

2009 - 20102

Settings

search.filters.applied.f.publicationcategory: search.filters.publicationcategory.Makale - Uluslararası Hakemli DergiSubject: search.filters.subject.simulated annealing

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Design Of Reliable Communication Networks: A Hybrid Ant Colony Optimization Algorithm
    (2010) Dengiz, Berna; Altiparmak, Fulya; Belgin, Onder; 0000-0003-1730-4214; 0000-0001-6702-2608; AAF-7020-2021; K-1080-2019
    This article proposes a hybrid approach based on Ant Colony Optimization (ACO) and Simulated Annealing (SA), called ACO_SA, for the design of communication networks. The design problem is to find the optimal network topology for which the total cost is a minimum and the all-terminal reliability is not less than a given level of reliability. The proposed ACO_SA has the advantages of the ability to find higher performance solutions, created by the ACO, and the ability to jump out of local minima to find better solutions, created by the SA. The effectiveness of ACO_SA is investigated by comparing its results with those obtained by individual application of SA and ACO, which are basic forms of ACO_SA, two different genetic algorithms and a probabilistic solution discovery algorithm given in the literature for the design problem. Computational results show that ACO_SA has a better performance than its basic forms and the investigated heuristic approaches.
  • No Thumbnail Available
    Item
    A Tabu Search Algorithm for the Training of Neural Networks
    (2009) Dengiz, B.; Alabas-Uslu, C.; Dengiz, O.
    The most widely used training algorithm of neural networks (NNs) is back propagation ( BP), a gradient-based technique that requires significant computational effort. Metaheuristic search techniques such as genetic algorithms, tabu search (TS) and simulated annealing have been recently used to cope with major shortcomings of BP such as the tendency to converge to a local optimal and a slow convergence rate. In this paper, an efficient TS algorithm employing different strategies to provide a balance between intensification and diversification is proposed for the training of NNs. The proposed algorithm is compared with other metaheuristic techniques found in literature using published test problems, and found to outperform them in the majority of the test cases.